The DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.
How independent are “independent” effects? relative risk estimation when correlated exposures are measured imprecisely
Abstract
A relative risk estimate which relates an exposure to risk of disease will tend to be estimated too close to unity if that exposure is subject to random measurement error or intra-subject variability. "Independent" relative risk estimates, for the effect of one exposure after adjusting for confounding exposures, may be biased in either direction, depending on the amount of measurement imprecision in the exposure of interest and in the confounders. We describe two methods which estimate the bias in multivariate relative risk estimates due to the effect of measurement imprecision in one or more of the exposure variables in the model. Results from the two methods are compared in an example involving HDL cholesterol, triglycerides and coronary heart disease. In this example, the degree of bias in relative risk estimates is shown to be highly dependent on the amount of measurement imprecision ascribed to the exposures. It is concluded that when two exposures are substantially correlated, and one or both is subject to sizeable measurement imprecision, a study in which exposures are measured only once will be inadequate for investigating the independent effect of the exposures. Where feasible, epidemiologists should seek study populations where the correlation between the exposures is smaller.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“However, when correlated risk factors are measured with different degrees of imprecision, the adjusted relative risk associated with them can be biased towards or away from unity [ 204 – 206 ].”
From Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration (2007) · cites this paper as [204]
Citations
Cites
Statistical Methods in Cancer Research
Breslow, N. E. (1980) · Cited by 7,332
(1977) · CERN Document Server (European Organization for Nuclear Research) · Cited by 1,548
Smoking and Lung Cancer: Recent Evidence and a Discussion of Some Questions
Cornfield, J., Haenszel, W. M., Hammond, E., Lilienfeld, A. M., Shimkin, M. B., & Wynder, E. L. (1959) · JNCI Journal of the National Cancer Institute · doi:10.1093/jnci/22.1.173 · Cited by 850
Hulley, S. B., Rosenman, R. H., Bawol, R. D., & Brand, R. J. (1980) · New England Journal of Medicine · doi:10.1056/nejm198006193022503 · Cited by 702
ROSNER, B. A., Willett, W. C., & Spiegelman, D. L. (1989) · Statistics in Medicine · doi:10.1002/sim.4780080905 · Cited by 658
British Regional Heart Study: cardiovascular risk factors in middle-aged men in 24 towns.
Shaper, A. G., Pocock, S., Walker, M., Cohen, N. M., Wale, C. J., & Thomson, A. G. (1981) · BMJ · doi:10.1136/bmj.283.6285.179 · Cited by 526
THE EFFECT OF MISCLASSIFICATION IN THE PRESENCE OF COVARIATES
GREENLAND, S. (1980) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a113025 · Cited by 494
ROSNER, B. A., Spiegelman, D. L., & Willett, W. C. (1990) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a115715 · Cited by 464
PLASMA TRIGLYCERIDE AS A RISK FACTOR FOR CORONARY HEART DISEASE
Austin, M. A. (1989) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a115130 · Cited by 351
Cited by
Sorted by citations
Items 201–275 of 275
Triglyceride and Cardiovascular Disease
Kim, M. (2013) · Journal of Lipid and Atherosclerosis · doi:10.12997/jla.2013.2.1.1 · Cited by 11
Commentary: Understanding pathways to cannabis use and from use to harm
Macleod, J., & Hickman, M. (2006) · International Journal of Epidemiology · doi:10.1093/ije/dyl104 · Cited by 11
Salenius, K., Väljä, N., Thusberg, S., Iris, F., Ladd‐Acosta, C., Roos, C., Nykter, M., Fasano, A., Autio, R., Lin, J., et al. (2024) · BMC Psychiatry · doi:10.1186/s12888-024-06392-w · Cited by 11
Smith, G. D., & Ebrahim, S. B. J. (2024) · International Journal of Epidemiology · doi:10.1093/ije/dyae179 · Cited by 10
Self-citationMacleod, J. (2004) · American Journal of Epidemiology · doi:10.1093/aje/kwh320 · Cited by 10
“The fact that the association changed little after adjustment for current income, education, and occupational class may largely reflect the limitations of statistical adjustment in cases where the adjustment factors are poor indices of the etiologically relevant confounders (6).”
BackgroundPhenome-wide Mendelian randomization study of plasma triglyceride levels and 2600 disease traits
Park, J. K., Bafna, S., Forrest, I. S., Duffy, Á., Márquez‐Luna, C., Petrazzini, B. O., Vy, H. M. T., Jordan, D. M., Verbanck, M., Narula, J., et al. (2023) · eLife · doi:10.7554/elife.80560 · Cited by 9
“They also correlate with LDL particles or apolipoprotein B (apoB) concentration ( Cromwell et al., 2007 ) and inversely correlated with high-density lipoprotein cholesterol (HDL-C) ( Phillips and Smith, 1991 ).”
Grant, A. D., & Wolf, G. (2019) · Journal of Circadian Rhythms · doi:10.5334/jcr.178 · Cited by 9
Padberg, I., Schneider, A., Rohmann, J. L., Kelley, S., Grittner, U., & Siegerink, B. (2020) · Respiratory Research · doi:10.1186/s12931-020-1305-6 · Cited by 9
Johansen, M. Ø., Moreno‐Vedia, J., Balling, M., Smith, G. D., & Nordestgaard, B. G. (2023) · Atherosclerosis · doi:10.1016/j.atherosclerosis.2023.117316 · Cited by 10
Self-citationWhy, when and how should hypertriglyceridemia be treated in the high-risk cardiovascular patient?
Watts, G. F., & Karpe, F. (2011) · Expert Review of Cardiovascular Therapy · doi:10.1586/erc.11.61 · Cited by 10
Is Acculturation in Hispanic Health Research a Flawed Concept
Ponce, C., & Comer, B. (2003) · Cited by 9
Measurement imprecision: ignore or investigate?
(1992) · The Lancet · doi:10.1016/0140-6736(92)90872-z · Cited by 9
Wade, K. H., Carslake, D., Nilsen, T. I. L., Timpson, N. J., Smith, G. D., & Romundstad, P. (2015) · Scientific Reports · doi:10.1038/srep12399 · Cited by 8
Self-citationKuang, S., Ma, X., Sun, L., Wang, C., Yang, L., Wang, G., Sun, J., Zhou, F., & Zhang, C. (2025) · ImmunoTargets and Therapy · doi:10.2147/itt.s510102 · Cited by 8
Smith, G. D., Albert Hofman, D., & Brennan, P. (2023) · European Journal of Epidemiology · doi:10.1007/s10654-023-01090-5 · Cited by 7
“The degree to which residual confounding can generate such findings has perhaps been under-appreciated, despite demonstrations of its potential to leave apparently convincing but spurious associations after statistical adjustment [49–52]”
Self-citationDepression and mortality following myocardial infarction: the issue of disease severity
Carroll, D., & Lane, D. A. (2002) · Epidemiologia e Psichiatria Sociale · doi:10.1017/s1121189x00005510 · Cited by 7
Personal paper: Gene therapy for cancer—managing expectations
Melcher, A. A., Garcia‐Ribas, I., & Vile, R. G. (1997) · BMJ · doi:10.1136/bmj.315.7122.1604 · Cited by 7
Macleod, J., Smith, G. D., & Lynch, J. W. (2004) · American Journal of Epidemiology · doi:10.1093/aje/kwh157 · Cited by 7
Self-citationCommentary: What should we make of associations between vital exhaustion and heart disease?
Macleod, J., & Carroll, D. (2003) · International Journal of Epidemiology · doi:10.1093/ije/dyg303 · Cited by 7
Cannabis use and psychosis: the origins and implications of an association
Macleod, J. (2007) · Advances in Psychiatric Treatment · doi:10.1192/apt.bp.106.003434 · Cited by 6
Homocysteine as a Risk Factor for Cerebrovascular Disease and Stroke
Perry, I. J. (2000) · Developments in cardiovascular medicine · doi:10.1007/978-94-017-1789-2_10 · Cited by 6
Lipid-lowering drug therapy: more knowledge leads to more problems for composers of guidelines
Durrington, P. N., & Illingworth, D. R. (2000) · Current Opinion in Lipidology · doi:10.1097/00041433-200008000-00001 · Cited by 6
Gustavson, K., Smith, G. D., & Eilertsen, E. M. (2022) · European Journal of Epidemiology · doi:10.1007/s10654-022-00857-6 · Cited by 5
Self-citationAn Epidemiological Study of Lung Cancer and Selected Other Cancers among Namibian Uranium Workers
Agius, R. M., Batistatou, E., Gittins, M., Jones, S. R., McNamee, R., Liu, H., Rashid, A., van Tongeren, M., von Oertzen, G., & Wakeford, R. (2023) · Radiation Research · doi:10.1667/rade-23-00051.1 · Cited by 5
“A factor that is not causally related to a disease, but measured with relatively high precision, can appear to be more important than a true causal risk factor that is measured with lower precision, if these factors are strongly correlated ( 35 ).”
Yarmolinsky, J., Lee, M. A., Lau, E., Moratalla-Navarro, F., Vincent, E. E., Li‐Gao, R., Rensen, P. C. N., van Dijk, K. W., Tsilidis, K. K., Sangphukieo, A., et al. (2025) · JNCI Journal of the National Cancer Institute · doi:10.1093/jnci/djaf137 · Cited by 5
Hickman, M., Macleod, J., & Smith, G. D. (2004) · Addiction · doi:10.1111/j.1360-0443.2004.00731.x · Cited by 5
Self-citationRegression adjustment for causal inference
Ho, F. K. W. (2025) · BMJ Medicine · doi:10.1136/bmjmed-2023-000816 · Cited by 5
“However, removing confounding bias based on regression adjustment relies on various untestable assumptions, such as accurately identifying and measuring confounders 6 and correctly modelling confounder-outcome associations.”
HIV-1 Viremia and Cancer Risk in 2.8 Million People: The South African HIV Cancer Match Study
Ruffieux, Y., Mwansa-Kambafwile, J. R. M., Metekoua, C., Tombe-Nyahuma, T., Bohlius, J., Muchengeti, M. M., Egger, M., & Rohner, E. (2024) · Clinical Infectious Diseases · doi:10.1093/cid/ciae652 · Cited by 4
“Differential measurement precision may, thus, have led to an overestimation of the association between HIV-1 viremia and cancer risk in the fully adjusted model [19].”
Zhang, H., & Zhou, Z. (2022) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2022.847583 · Cited by 5
“However, the results of conventional observational studies may have been affected by residual confounding and reverse causation (Phillips and Smith, 1991), thus limiting our knowledge of the effect fibrinogen on the risk of NDs. Mendelian randomization (MR) is a method that uses genetic variants,…”
BackgroundMethodMeasurement Errors in Epidemiology
Greenwood, D. C. (2012) · Springer eBooks · doi:10.1007/978-94-007-3024-3_3 · Cited by 4
Statistical correction for measurement imprecision
Salonen, J. (1992) · The Lancet · doi:10.1016/0140-6736(92)90711-b · Cited by 4
Commentary: Triglycerides or HDL cholesterol in cardiovascular disease—which is the true culprit?
Varbo, A., & Nordestgaard, B. G. (2019) · International Journal of Epidemiology · doi:10.1093/ije/dyy292 · Cited by 4
Amoroso, T. (2018) · Drug Science Policy and Law · doi:10.1177/2050324518789214 · Cited by 3
“Another reason why meta-analyses of observational research can be misleading is that confounding variables cannot be precisely controlled for in the original studies they include (Phillips and Smith, 1991; Smith and Phillips, 1992).”
BackgroundWhen and how to measure lipids and their role in CHD risk prediction
Richmond, W. (2003) · The British Journal of Diabetes · doi:10.1177/14746514030030030601 · Cited by 3
Health promotion in general practice. Room for improvement in data collection.
Wilson, A., Dowell, T., Pollock, C., & Weekes, T. (1993) · BMJ · doi:10.1136/bmj.307.6900.381 · Cited by 3
Understanding lipidomics associations and the lipoprotein-related caveats in population epidemiology
Zhao, S., Ohukainen, P., Kettunen, J., Järvelin, M., Kähönen, M. A. P., Lehtimäki, T. J., Viikari, J., Raitakari, O. T., Mäkinen, V., & Ala‐Korpela, M. (2024) · American Journal of Epidemiology · doi:10.1093/aje/kwae445 · Cited by 2
Genetic and Nutritional Interactions in Cardiovascular Disease
Caterina, R. D. (2011) · World review of nutrition and dietetics · doi:10.1159/000327803 · Cited by 2
Næss, Ø. (2009) · Norsk Epidemiologi · doi:10.5324/nje.v15i1.230 · Cited by 2
“Until we know better how these social and biological factors are interwoven and operate along the life course, it will be misleading to claim they have an independent effect (42,43).”
BackgroundDoes occupational success influence longevity among England test cricketers?
Boyle, P. J. (2008) · British Journal of Sports Medicine · doi:10.1136/bjsm.2007.041566 · Cited by 2
“Teasing out the occupational determinants of survival—whether material, success or control factors—is not a simple task as health outcomes may relate to complex relationships between command over resources, perceptions of social hierarchy and other factors which influence general material inequality.”
BackgroundSmith, G. D., Hart, C., Blane, D., & Hole, D. J. (2003) · Policy Press eBooks · doi:10.56687/9781447342229-020 · Cited by 2
Self-citationPulmonary dysfunction and development of different cardiovascular outcomes in the general population
Padberg, I., Schneider, A., Grittner, U., Olma, M. C., Liman, T., & Siegerink, B. (2017) · Archives of cardiovascular diseases · doi:10.1016/j.acvd.2017.07.001 · Cited by 2
Turrell, G., Blakely, T. A., Patterson, C. M., & Oldenburg, B. F. (2004) · Journal of Epidemiology & Community Health · doi:10.1136/jech.2003.011031 · Cited by 114
“[54][55[56] Fourth, the inclusion of individual-level covariates in multilevel analyses may result in over-control, which argues for the possibility of a true contextual-effect on food”
BackgroundBias due to measurement imprecision
Phillips, A., & Smith, G. D. (1992) · The Lancet · doi:10.1016/0140-6736(92)91242-z · Cited by 2
Self-citationO’Seaghdha, C. M., & Fox, C. S. (2011) · Journal of the American Society of Nephrology · doi:10.1681/asn.2011050488 · Cited by 2
Pitfalls in epidemiological analysis
Carlsson, G. (1997) · Scandinavian Journal of Social Medicine · doi:10.1177/140349489702500203 · Cited by 2
“voices have been raised, more recently by Phillips and Smith and in a Lancet editorial (5, 6).”
BackgroundNeurocysticercosis: households, pigs, and risks
Miranda, J. J. (2003) · Journal of Epidemiology & Community Health · doi:10.1136/jech.57.8.570 · Cited by 2
HIV Infection and Smoking Behavior
Smith, G. D. (1992) · JAMA · doi:10.1001/jama.1992.03490120053020 · Cited by 2
Self-citationTechnical Variability in Laboratory Data
Vineis, P., Schulte, P. A., & Vogt, R. F. (1993) · Elsevier eBooks · doi:10.1016/b978-0-12-632345-0.50008-0 · Cited by 2
Cardiovascular complications of non-insulin-dependent diabetes. The difficult search for causality.
Fontbonne, A. (1993) · Circulation · doi:10.1161/01.cir.88.4.1952 · Cited by 1
Determinants of the East-West Life Expectancy Gap: Contribution of Lifestyle and Behaviour
Bobák, M. (1996) · Springer eBooks · doi:10.1007/978-94-009-1764-4_7 · Cited by 1
Pan, H., Hu, J., Gong, S., & Fei, Y. (2025) · Discover Oncology · doi:10.1007/s12672-025-02704-y · Cited by 1
“MR emerges as a potent causal inference method, leveraging multiple genetic variants as instrumental variables to test specific hypotheses [24, 27].”
Health promotion in general practice. Simple education is never enough.
Sheldon, T., Smith, G. D., & Davison, C. (1993) · BMJ · doi:10.1136/bmj.307.6900.380-c · Cited by 1
Self-citationRenal replacement treatment. Elderly people must not be denied CAPD.
Solanki, T., & Lambert, L. A. (1993) · BMJ · doi:10.1136/bmj.307.6900.381-a · Cited by 1
ADIPOCYTE-DERIVED HORMONES AND CARDIOVASCULAR DISEASE
Eriksson, M. (2010) · DiVA at Umeå University (Umeå University) · Cited by 1
Commentary: Chylomicronaemia, fat tolerance and atherosclerosis—a commentary on a landmark paper
Karpe, F. (2016) · International Journal of Epidemiology · doi:10.1093/ije/dyw214 · Cited by 1
Xue, X., Chen, B., Jiang, X., Jin, C., Yue, M., Li, Z., Wang, Q., & Zhang, L. (2023) · Research Square · doi:10.21203/rs.3.rs-3496935/v1 · Cited by 1
Collinearity and imprecision in dietary measurement
Phillips, A. N. (1992) · Journal of Clinical Epidemiology · doi:10.1016/0895-4356(92)90141-9
Self-citationRenal replacement treatment. CAPD has its limitations.
Tattersall, J. E., Greenwood, R. N., & Earrington, K. (1993) · BMJ · doi:10.1136/bmj.307.6900.381-b
Dentine lead and intelligence prior to school entry: a statistical sensitivity analysis
Grandjean, P. (1993) · Journal of Clinical Epidemiology · doi:10.1016/0895-4356(93)90155-t
Commentary: Cheap but cheerless. Is it the underpaid or the overworked who die young?
Metcalfe, C. D. (2002) · International Journal of Epidemiology · doi:10.1093/ije/31.3.622
Abímbọ́lá, Ṣ. (2010) · UPT. Syiah Kuala University Library (Syiah Kuala University)
Telles‐Correia, D., Barbosa, A., & Mega, I. (2010) · DOAJ (DOAJ: Directory of Open Access Journals) · doi:10.20344/amp.690
mental health? What is the association between wealth and
Carter, K. N., Blakely, T. A., Collings, S. C., Gunasekara, F. I., & Richardson, K. (2009)
between married and unmarried men
Ben‐Shlomo, Y., Smith, G. D., Shipley, M. J., & Marmot, M. G. (1993)
Self-citationLawford‐Smith, H. (2015)
Hartwig, F. P., Borges, M. C., Horta, B. L., Bowden, J., & Smith, G. D. (2017) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/123976
“Conventional observational studies may have important limitations such as reverse causation and residual confounding, 17-19 which hamper .”
BackgroundSelf-citationGreene, T., & Ernhart, C. B. (1993) · Journal of Clinical Epidemiology · doi:10.1016/0895-4356(93)90156-u
Discussion on the Paper by Coxet al.
(1992) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/j.1467-985x.1992.tb00382.x
Smith, G. D., Hart, C., Blane, D., & Hole, D. J. (2003) · Policy Press eBooks · doi:10.51952/9781447342229.ch017 · Cited by 2
Self-citationLi, B., Zou, L., Chen, M., Liao, M., Gong, Z., Cheng, R., Ma, Z., Zheng, Q., & Wang, H. (2023) · Research Square · doi:10.21203/rs.3.rs-2822776/v1
Salenius, K., Väljä, N., Thusberg, S., Iris, F., Ladd‐Acosta, C., Roos, C., Nykter, M., Fasano, A., Autio, R., & Lin, J. (2024) · medRxiv · doi:10.1101/2024.01.07.24300940
“We conducted Mendelian Randomization analyses (MR) on the selected autism multi-trait SNPs based on their assigned central traits, to explore the liability and independent (reverse causation) relationships between autism and its related biomarkers (Phillips and Smith 1991).”
MethodXiao, Y., Li, X., Zou, L., Wall, D. M., & Liang, Z. (2024) · medRxiv · doi:10.1101/2024.04.23.24306212
“However, a causal role for the gut microbiota in PC has not been established, due to confounding factors and the poten/al for reverse causality [14-16].”
BackgroundKong, J., Liu, X., Li, H., Yang, C., Jiang, T., Yan, Y., Miao, N., Mu, S., & Zhan, Y. (2025) · Cytokine · doi:10.1016/j.cyto.2024.156849
Adiposity measures and risk of cardiovascular disease
Wormser, D. (2012) · Apollo (University of Cambridge) · doi:10.17863/cam.16249
“AMORIS Sweden 1987-1991 Screening Complete Assessed 58082 25 (4) 46 (10) 33334 (57) ARIC USA 1987-189 Households Random Assessed 14600 28 (5) 54 (6) 6302 (43) ATENA Italy 1994-1996 General population Random Assessed 4741 27 (4) 50 (7) 0 (0) ATS_SAR Italy 1983-1984 Combination or other Complete & random Assessed 4263 2…”
Yarmolinsky, J., Cavallo, F. R., Koskeridis, F., Yu, X., Bouras, E., Richenberg, G., Costantini, I., Ray, D., Woolf, B., Karhunen, V., et al. (2026) · medRxiv · doi:10.64898/2026.08.30.26361622
Self-citation
Metadata
- Authors
- Phillips, A. N., & Smith, G. D.
- Affiliations
- The Royal Free Hospital (GB)
- London School of Hygiene & Tropical Medicine (GB)
- Corresponding author
- Phillips, Andrew N.
- Venue
- Journal of Clinical Epidemiology
- Publisher
- Elsevier BV
- Published
- 1991-01
- Volume
- 44
- Issue
- 11
- Pages
- 1223-1231
- ISSN
- 0895-4356 (ISSN-L)
- OpenAlex
- W2031403388
- Cited by
- 275
- References
- 35
- Citations by year
- 2026: 1 · 2025: 5 · 2024: 12 · 2023: 9 · 2022: 8 · 2021: 3 · 2020: 10 · 2019: 8 · 2018: 3 · 2017: 8 · 2016: 8 · 2015: 7 · 2014: 2 · 2013: 5 · 2012: 5
- FWCI
- 22.12
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Journal of Clinical Epidemiology (publishedVersion)
- PubMed (publishedVersion)
- UCL Discovery (University College London) (submittedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Keywords
- relative risk estimation, measurement error, confounding, HDL cholesterol, triglycerides, coronary artery disease
- MeSH terms
- Humans, Cholesterol, HDL, Mathematical Computing, Myocardial Infarction, Probability, Risk, Triglycerides, Bias, Confounding Factors, Epidemiologic, Multivariate Analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
